Active vibration isolation with neural network for ship propulsion shafting
                                                                            XU Hui, ZHU Changsheng
                                    (College of Electrical Engineering, Zhejiang University, Hangzhou 310027, China)

Abstract: Aiming at the problem that the multifrequency forces caused by the propeller transmitted to the hull through the propulsion shafting, which has a series of bad effects on the ship, in order to suppress the transmission forces, a method of multifrequency forces active vibration isolation was proposed, which was achieved by installing actuators at the bearing base. Firstly, the influence on the whole shafting and the feasibility of the proposed method were analyzed by the established dynamic model of the propulsion shafting. Then, considering the influence of error channel, a neural network controller with error channel identification module was proposed. The weights of the optimal control force for the shafting were iterated by the error signal. Finally, the shafting model and simulation platform were built in Matlab/Simulink. The results indicate that the transmission force can be effectively isolated by using the proposed method.

Key words: propulsion shafting; active vibration isolation; finite element; neural network; identification module